European Patent 309,477 relates to ecteinascidins 729, 743, 745, 759A. 759B and 770. The ecteinascidin compounds are disclosed to have antibacterial and other useful properties. Ecteinascidin 743 is now undergoing clinical trials as an antitumour agent.
Ecteinascidin 743 has a complex tris(tetrahydroisoquinolinephenol) structure of the following formula (I):

In ecteinascidin 743, the 1,4 bridge has the structure of formula (IV):

Other known ecteinascidins include compounds with a different bridged cyclic ring system such as occurs in ecteinascidin 722 and 736, where the bridge has the structure of formula (V):
ecteinascidins 583 and 597, where the bridge has the structure of formula (VI):
and ecteinascidin 594 and 596, where the bridge has the structure of formula (VII):

The complete structure for these and related compounds is given in J. Am. Chem. Soc. (1996) 118, 9017–9023. This article is incorporated by reference.
The ecteinascidins are currently prepared by isolation from extracts of the marine tunicate Ecteinascidiii turbinata. The yield is low, and alternative preparative processes have been sought.
A synthetic process for producing ecteinascidin compounds is described in U.S. Pat. No. 5,721,362, see also WO 9812198. The claimed method is long and complicated. By way of illustration, there are 38 Examples each describing one or more steps in the synthetic sequence to arrive at ecteinascidin 743.
Claim 25 of U.S. Pat. No. 5,721,362 is directed at an intermediate phenol compound of a given formula (11), which we refer to also as Intermediate 11 or Int-11. It has the following bis(tetrahydroisoquinolinephenol) structure (II):
where MOM is a methoxymethyl substituent and TBDPS is a tert-butyldiphenylsilyl substituent.
From Intermediate 11 it is possible to synthesise another interesting antitumour agent, phthalascidin, see Proc. Natl. Acad. Sci. USA, 96, 3496–3501, 1999. Phthalascidin is a bis(tetrahydroisoquinolinephenol) derivative of formula (III):

More generally, phthalascidin and related compounds are described in WO 0018233. Claim 1 is directed at compounds of formula:
    wherein the substituent groups defined by R1, R2, R3, R5, R6, R7, R8 and R9 are each independently selected from the group consisting of H, OH, OR′, SH, SR′, SOR′, SO2R′, NO2, NH2, NHR′, N(R′)2, NHC(O)R′, CN, halogen, ═O, C(═O)H, C(═O)R′, CO2H, CO2R′, C1–C12 alkyl, C2–C12 alkenyl, C2–C12 alkynyl, substituted or unsubstituted aryl, substituted or unsubstituted aralkyl, and substituted or unsubstituted heteroaromatic;    wherein each of the R′ groups is independently selected from the group consisting of H, OH, NO2, NH2, SH, CN, halogen, ═O, C(═O)H, C(═O)CH3, CO2H, CO2CH3, C1–C12 alkyl, C2–C12 alkenyl, C2–C12 alkynyl, aryl, aralkyl, and heteroaromatic;    wherein each dotted circle represents one, two or three optional double bonds;    wherein R7 and R8 may be joined into a carbocyclic or heterocyclic ring system;    and wherein X1 and X2 are each independently defined as above for R1–R8 and further include various permitted definitions.
Further naturally occuring compounds are known which lack a bridged cyclic ring system. They include the bis(tetrahydroisoquinolinequinone) antitumor-antimicrobial antibiotics safracins and saframycins and the marine natural products renieramicins and xestomycin isolated from cultured microbes or sponges. They all have a common dimeric tetrahydroisoquinoline carbon framework. These compounds can be classified into four types, types I to IV, with respect to the oxidation pattern of the aromatic rings.
Type I, dimeric isoquinolinequinones, is a system of formula (VIII) most commonly occurring in this class of compounds, see the following table I.
TABLE IStructure of Type I Saframycin Antibiotics. SubstituentsCompoundR14aR14bR21R25aR25bR25csaframycin AHHCNOOCH3saframycin BHHHOOCH3saframycin CHOCH3HOOCH3saframycin GHOHCNOOCH3saframycin HHHCNOHCH2COCH3CH3saframycin SHHOHOOCH3saframycin Y3HHCNNH2HCH3saframycin Yd1HHCNNH2HC2H5saframycin Ad1HHCNOOC2H5saframycin Yd2HHCNNH2HHsaframycin Y2bHQbCNNH2HCH3saframycin Y2b−dHQbCNNH2HC2H5saframycin AH2HHCNHaOHaCH3saframycin AH2AcHHCNHOAcCH3saframycin AH1HHCNOHaHaCH3saframycin AH1AcHHCNOAcHCH3saframycin AR3HHHHOHCH3aassignments are interchangeable.bwhere the group Q is of formula (IX):
Type I aromatic rings are seen in saframycins A, B and C; G and H; and S isolated from Streptomyces lavendulae as minor components. A cyano derivative of saframycin A, called cyanoquinonamine, is known from Japanese Kokai JP-A2 59/225189 and 60/084,288. Saframycins Y3, Yd1, Ad1, and Yd2 were produced by S. lavenidulae by directed biosynthesis, with appropriate supplementation of the culture medium. Saframycins Y2b and Y2b-d dimers formed by linking the nitrogen on the C-25 of one unit to the C-14 of the other, have also been produced in supplemented culture media of S. lavendulae. Saframycins AR, (=AH2,), a microbial reduction product of saframycin A at C-25 produced by Rhodococcus amidophilus, is also prepared by nonstereoselective chemical reduction of saframycin A by sodium borohydride as a 1:1 mixture of epimers followed by chromatographic separation [the other isomer AH1 is less polar]. The further reduction product saframycin AR3, 21-decyano-25-dihydro-saframycin A. (=25-dihydrosaframycin B) was produced by the same microbial conversion. Another type of microbial conversion of saframycin A using a Nocardia species produced saframycin B and further reduction by a Mycobacterium species produced saframycin AH1Ac. The 25-O-acetates of saframycin AH2 and AH1 have also been prepared chemically for biological studies.
Type I compounds of formula (X) have also been isolated from marines sponges, see Table II.
TABLE IIStructures of Type I Compounds from Marine Sponges. SubstituentsR14aR14bR21Rrenieramycin AOHHH—C(CH3)═CH—CH3renieramycin BOC2H5HH—C(CH3)═CH—CH3renieramycin COHOO—C(CH3)═CH—CH3renieramycin DOC2H5OO—C(CH3)═CH—CH3renieramycin EHHOH—C(CH3)═CH—CH3renieramycin FOCH3HOH—C(CH3)═CH—CH3xestomycinOCH3HH—CH3
Renieramycins A–D were isolated from the antimicrobial extract of a sponge, a Reniera species collected in Mexico, along with the biogenetically related monomeric isoquinolines renierone and related compounds. The structure of renieramycin A was initially assigned with inverted stereochemistry at C-3, C-11, and C-13. However, careful examination of the 1H NMR data for new, related compounds renieramycins E and F, isolated from the same sponge collected in Palau, revealed that the ring junction of renieramycins was identical to that of saframycins. This result led to the conclusion that the formerly assigned stereochemistry of renieramycins A to D must be the same as that of saframycins.
Xestomycin was found in a sponge, a Xestosponigia species collected from Sri Lancan waters.
Type II compounds of formula (XI) with a reduced hydroquinone ring include saframycins D and F, isolated from S. lavendulae, and saframycins Mx-1 and Mx-2, isolated from Myxococcus xanthlus. See table III.
TABLE IIIType II Compounds SubstituentsCompoundR14aR14bR21R25aR25aR25csaframycin DOOHOOCH3saframycin FOOCNOOCH3saframycin Mx-1HOCH3OHHCH3NH2saframycin Mx-2HOCH3HHCH3NH2
The type III skeleton is found in the antibiotics safracins A and B, isolated from cultured Pseudomonas fluorescens. These antibiotics of formula (XII) consist of a tetrahydroisoquinoline-quinone subunit and a tetrahydroisoquninolinephenol subunit.
where R21 is —H in safracin A and is —OH in safracin B.
Saframycin R, the only compound classified as the Type IV skeleton, was also isolated from S. lavendulae. This compound of formula (XIII), consisting of a hydroquinone ring with a glycolic ester sidechain on one of the phenolic oxygens, is conceivably a pro-drug of saframycin A because of its moderate toxicity.

All these known compounds have a fused system of five rings (A) to (E) as shown in the following structure of formula (XIV):

The rings A and E are phenolic in the ecteinascidins and some other compounds, while in other compounds, notably the saframycins, the rings A and E are quinolic. In the known compounds, the rings B and D are tetrahydro, while ring C is perhydro.